Taiyi Wang — accepted AI-conference papers

2026

OServe: Accelerating LLM Serving via Spatial-Temporal Workload Orchestration

Youhe Jiang, Fangcheng Fu, Taiyi Wang, Guoliang HE, Eiko Yoneki

ICML 2026poster

Serving Large Language Models (LLMs) can benefit immensely from parallelizing both the model and input requests across multiple devices, but incoming workloads exhibit substantial spatial and temporal heterogeneity. Spatially, workloads comprise heterogeneous requests with varying compute and memory…

Cited by 0 Source ↗Scholar ↗

2025

DistRL: An Asynchronous Distributed Reinforcement Learning Framework for On-Device Control Agent

Taiyi Wang, Zhihao Wu, Jianheng Liu, Jianye HAO, Jun Wang, Kun Shao

ICLR 2025poster

On-device control agents, especially on mobile devices, are responsible for operating mobile devices to fulfill users' requests, enabling seamless and intuitive interactions. Integrating Multimodal Large Language Models (MLLMs) into these agents enhances their ability to understand and execute compl…

Cited by 13 Source ↗PDF ↗Scholar ↗